1. Field of the Invention
The present invention relates to a technique for registering object data included in an image which is read from an original (document), and searching for a desired object in the registered object data.
2. Description of the Related Art
Hitherto, a technique for converting an input image to vector data has been practiced in a document processing system to facilitate reutilization of information that has been stored in the form of paper documents and image data (see Japanese Patent Laid-Open No. 2004-326490). With the technique disclosed in Japanese Patent Laid-Open No. 2004-326490, an image having been read by a scanner and stored is converted to vector data so that a file search can be performed and convenience in reutilization can be improved.
It is not always assured that any image having been read by a scanner and stored as vector data is convenient for a user to utilize data of the stored image. From this point of view, the technique disclosed in Japanese Patent Laid-Open No. 2004-326490 still has a room for improvement.
For example, when a catalogue explaining a new car is scan images read by the scanner include not only an image of the car itself, but also various kinds of information, such as explanation texts. After the scanning, when a user makes a search of the read car-catalogue images in preparing a new catalogue for another car, the input car-catalogue images including not only the image of the car itself, but also various kinds of information, such as explanation texts, are all searched.
In the above-described situation, if only the car image that is data as a part of the input catalogue images (i.e., object data) can be extracted by the search of the read car-catalogue image, the user can obtain image data in the form more convenient to reutilize the same. As another case, when the user prepares a document by utilizing the object data for a different purpose, the user may want to obtain combined information of both the car image and the explanation texts, for example, as the search result.
Stated another way, data that is optimum for the user making a search of the car catalogue images to utilize a part of those images again differs depending on the user's intention. Namely, of the input car-catalogue images, the user wants to obtain only the car image in some cases and the combined information of both the car image and the explanation texts in other cases.